deboraJ23 commited on
Commit
c4669d1
·
verified ·
1 Parent(s): 9d19cc8

Create README.md

Browse files
Files changed (1) hide show
  1. README.md +82 -0
README.md ADDED
@@ -0,0 +1,82 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ datasets:
3
+ - ODELIA-AI/ODELIA-Challenge-2025
4
+ language:
5
+ - en
6
+ metrics:
7
+ - roc_auc
8
+ pipeline_tag: image-classification
9
+ tags:
10
+ - breast
11
+ - cancer
12
+ - mri
13
+ - odelia
14
+ - resnet
15
+ - 3d-classification
16
+ extra_gated_prompt: >-
17
+ ### Model Usage Agreement
18
+
19
+ By accessing or using this model (the "Model"), you acknowledge and agree
20
+ to the following terms:
21
+
22
+ 1. The Model is provided for non-commercial academic and research use only.
23
+ 2. The Model must not be used for diagnosis, treatment, clinical decision
24
+ making, or any application involving real-patient care.
25
+ 3. Do not identify, re-identify, or deanonymize people whose data may have
26
+ contributed to the training or evaluation data.
27
+ 4. Publications or derivative research must acknowledge the ODELIA
28
+ consortium and cite the Pimed authors and source repository.
29
+ 5. You are responsible for validating outputs in your research context.
30
+ The Model is provided as-is, without warranties.
31
+
32
+ This gated access does not grant rights beyond the license and permissions
33
+ stated in this model card. Third-party components retain their own licenses.
34
+ ---
35
+
36
+ # ODELIA Pimed Model
37
+
38
+ Three-class 3D breast MRI classification model contributed by the Pimed team
39
+ to the ODELIA Challenge 2025.
40
+
41
+ The source repository is
42
+ [dnschouten/odelia_submission_pimed](https://github.com/dnschouten/odelia_submission_pimed).
43
+
44
+ ## Method
45
+
46
+ The corresponding implementation in the ODELIA project uses a 3D ResNet-18
47
+ backbone with one input channel and three output classes. It wraps the model
48
+ for subject-level classification of ODELIA MRI volumes.
49
+
50
+ The exact preprocessing, checkpoint configuration, and evaluation procedure
51
+ must be taken from the authors' source repository and the files uploaded with
52
+ this model. Do not infer clinical performance from the architecture alone.
53
+
54
+ ## Intended use and limitations
55
+
56
+ This release is for research and benchmarking only. It is not a medical
57
+ device, is not clinically validated, and must not be used for diagnosis,
58
+ prognosis, treatment, triage, or other clinical decisions. Performance may
59
+ vary across institutions, scanners, acquisition protocols, and populations.
60
+
61
+ ## Setup
62
+
63
+ Use the source repository for the authoritative setup and inference commands:
64
+
65
+ ```bash
66
+ git clone https://github.com/dnschouten/odelia_submission_pimed.git
67
+ cd odelia_submission_pimed
68
+ ```
69
+
70
+ ## License and attribution
71
+
72
+ The implementation uses PyTorch/MONAI components. Their applicable licenses,
73
+ notices, and any pretrained-weight terms must be preserved and reviewed before
74
+ redistribution.
75
+
76
+ Please cite or acknowledge:
77
+
78
+ - ODELIA Challenge 2025 and the ODELIA consortium.
79
+ - The Pimed authors: Daan Schouten, Jeong Hoon Lee, Mirabela Rusu and [source repository](https://github.com/dnschouten/odelia_submission_pimed).
80
+ - The ODELIA paper: [arXiv:2506.00474](https://arxiv.org/abs/2506.00474).
81
+
82
+ For questions regarding model architecture, training, or permissions, please contact **Daan Schouten**.